The proposed research will test the causal effects of social capital on levels and inequalities of children's social and cognitive development during the early elementary years. Social capital here refers to trust and shared expectations embedded in social networks of parents, teachers, and children. For young children, social capital operates primarily through their relationships with their parents, enhancing development through mechanisms of social support and social control. Children from low-income Latino families are a special focus of the study. The research design is experimental: social capital will be manipulated through a well-tested intervention Families and Schools Together (FAST) that will enhance the social capital among parents, teachers, and children through an intensive after-school program and a 2-year follow-up program. The study will include 52 schools in Milwaukee and San Antonio (about 208 teachers and 3,120 first graders and their families), with half the schools randomly selected for the intervention and half serving as controls. The experimental design will permit a strong test of the causal role of social capital. The project will also involve an implementation study intended to illuminate the processes by which social capital is generated and affects child development. The proposed research is interdisciplinary, drawing on sociology, psychology, and economics,and it reflects an ecological perspective that crosses boundaries between families and schools. The study thus responds to the call for research on the Science and Ecology of Early Development. Key aspects of child development will be assessed, including (a) social skills and problem behavior from standardized behavioral ratings by parents; (b) the latter plus academic competence as rated by teachers; and (c) grade retention, attendance rates, and third-grade reading and mathematics scores from school records. Social capital will be measured with repeated surveys of teachers and parents that address the extent of social networks, parent involvement, trust, and shared expectations among parents, between parents and schools, and between parents and children. Data will be analyzed with multilevel statistical models that take account of variation at the individual and school levels of analysis.RELEVANCE. Latinos are the fastest growing segment of the U.S. population, and among the most disadvantaged. Latino families often perceive a sense of distance from the school system, which can create barriers to their children's academic and social development, important dimensions of public health. This study will yield practical as well as scientific insights into the value of increased social capital for development of children among a variety of demographic groups, and particularly those from low-income Latino families.